From scenario-based seismic hazard to scenario-based landslide hazard: fast-forwarding to the future via statistical simulations

Abstract Ground motion scenarios exists for most of the seismically active areas around globe. They essentially correspond to shaking level maps at given earthquake return times which are used as reference likely under threat from future ground displacements. Being landslides in actively regions closely controlled by motion, one would expect that landslide susceptibility should change patterns space and time. However, so far, statistically-based assessments have primarily been time-invariant.In other words, vast majority statistical models does not include temporal effect main trigger scenarios. In this work, we present an approach aimed filling gap, bridging current practices seismological community those geomorphological ones. More specifically, select earthquake-induced inventory corresponding 1994 Northridge build a Bayesian Generalized Additive Model binomial family, featuring common morphometric thematic covariates well Peak Acceleration generated earthquake. Once each model component has estimated, run 1000 simulations 217 possible study area. From batch simulations, estimated mean 95% Credible Interval represent pattern specific scenario, together with its uncertainty level. Because scenario time, our allow incorporate dimension into any model, therefore driving results toward definition hazard. Ultimately, also share vector format – .mif file can be easily converted shapefile –. There, report (and uncertainty) simulation